.png){kind=icon}
.png){kind=icon}
Return
Enhanced Vessel Trajectory Prediction With Novel Mamdani Fuzzy Inference System-Based Alpha--Beta Filter Using AIS Data
.png){kind=icon}
PRE
AI
J
U
A
S
E
K
DOI：10.1109/tits.2026.3695566.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Shipping is integral to global trade, with increasing maritime traffic elevating safety risks. Accurate trajectory prediction based on Automatic Identification System (AIS) data is essential to prevent collisions and enhance navigational safety. This study introduces a novel Mamdani Fuzzy Inference System (MFIS)-based alpha-beta filter for vessel trajectory prediction, addressing the limitations of conventional static parameter tuning. The proposed method dynamically adjusts alpha and beta parameters using fuzzy logic and processes speed, heading, latitude, and longitude inputs to predict trajectories over extended periods. The fuzzy alpha-beta filter’s performance was evaluated on AIS data from 50 ships using metrics such as Average Displacement Error (ADE), Final Displacement Error (FDE), Non-Linear ADE (NL-ADE), and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R^{2}$ </tex-math></inline-formula>. It achieved superior accuracy (ADE: 0.020, FDE: 0.031, NL-ADE: 0.011, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R^{2}$ </tex-math></inline-formula>: 0.97) compared to conventional and deep learning models, demonstrating its efficacy in complex maritime scenarios. These findings mark a significant advancement in trajectory prediction systems, enhancing maritime safety and operational efficiency.
Keywords:
AIS data
Mamdani fuzzy inference system
vessel trajectory prediction
alpha-beta filter
fuzzy system
autonomous ship
Journal
IF:
8.4
Papers:
9.5K
Citations:
6.3W
Organization
Cited Papers